To explore the possibility of a new way to massively sequence DNA, we propose to decorate the inside of a DNA polymerase with one or more fluorescent reporter probes that are predicted to actually touch the template bases, according to several X-ray crystal structures. The crystallographic structures show some of the amino acid residues in very different positions and at very different angles during the reaction coordinate. Similarly, each template base appears to be flipped out by 90 degrees before it is swung back into base-pairing position, reportedly stacking on a certain amino acid residue. If we can observe single molecules of positionally fixed, processive, specially-fluorescent DNA polymerases as they synthesize DNA using normal dNTPs, we believe there will be some reproducible flickering, and that this flickering pattern might be empirically diagnostic of the identity of each template base in turn. Signal level (quenching or enhancement), wavelength (a shift of the peak of fluorescence) and/or a change in polarity are anticipated. Alternatively or additionally, if the spatial position and angle of the base- touching fluor is different for each template base, we expect useful alteration of the fluorescence of one or more FRET acceptor fluors positioned 10-15 angstroms away. If some form of our idea works, it could be relatively inexpensive, since no machined chips, no unusual dNTPs, and no PCR emulsions need be involved. (but would be allowed). A great deal of computational analysis of video (taken through a prism and/or a polarizing filter) should be involved. In one scenario, the subject DNA could be gapped genomic DNA, combed out, so that the analyzed DNA polymerase molecules at the gaps, lined up in a row, would indicate relative genomic map position of the sequences that they are flashing.